Data Strategy: Introduction

An introduction to the series "Data Strategy". A guide on how to becoming a data-driven organisation.

We often see large tech-companies making the news with amazing projects using data. These projects draw attention as they use the latest techniques or make for an interesting discussion at presentations, workshops or just at the water cooler.

We get inspired by the large tech-companies and see chances for improvement by implementing these new techniques in our own company as fast as possible. However, once started, you will notice that applying these techniques is harder than it seems. These tech-companies have enormous budgets, have little to no legacy technology, are often created from a digital mindset and have products and/or services that are well suited for the technique.

In this series, "Data Strategy", we want to help organisations to take the next step in the data landscape through sharing knowledge and best practices using our past experiences. The series will cover the following topics:

  • Education and knowledge sharing
  • Building an expert team
  • Creating a data platform
  • Applying machine learning

We will write about these five topics in the next blog posts of these series, the post of today will function as an introduction.

Creating the first movement

The five topics described in this series are essential for data-driven working and will provide the most value when executed in parallel. This will allow learned lessons to be shared immediately and provide value directly. Another advantage of executing those in parallel is the emergence of a movement within the organisation. Only by having the entire company work on this transition, real steps towards a data-driven organisation can be taken. With our inspiring sessions, we give this an extra boost.

Target Audience

This series is meant for organisations that want to start their transformation to become more data-driven, and organisations that are experiencing difficulties with this process. The next post in this series will show why education is essential to a team of data experts, and show what advantages are created by educating the entire organisation on the basic opportunities of data science.

Terms

Many terms used in data science have marketing buzzwords, which can cause some confusion, as these terms will have different meanings to different people. In order to clarify our story, we elaborate on what some of these terms mean to us as data scientists.

Data Science
At Gyver, we use the term Data Science to indicate to process which translates data to information. The components within this process are:

Business understanding
Determining and translating the needs of the user.

Data mining
Obtaining data from various sources.

Data cleaning / Data Preparation
Cleaning, preparing and combining data in order to translate this data to valuable information.

Data Exploration
Conducting exploring and question-driven analysis on the data.

Feature engineering / Predictive modelling
Creating algorithms that make predictions based on data.

Data visualisation: 
Visualising the information to give new insights to the end-user.

Artificial Intelligence:
Narrow AI: Technologies that can perform a certain specific task with human accuracy.
General AI: Machines that have a substitute to human senses, and can think and reason like humans.

Machine Learning:
A technique that can be used to achieve artificial intelligence. It consists of an algorithm that can interpret data, learn from this data and create a prediction of something in the real world.

Data Lake / Datahub
A central location for data storage and transformation.

Big data
We speak of "big data" when the data is of a certain volume, contains many different features and has a high renewal rate.

Cloud services / Cloud storage / Cloud computing
Traditionally, data is stored on one or multiple servers, requiring expensive purchase and maintenance, and setting a static storage capacity. Cloud providers, such as Google, Amazon or Microsoft, provide server capacity which can be accessed through the internet. In this way, storage capacity and computing power are dynamic.

Wrap-up

So these were the first concepts on our view of a Data Strategy. In our next blogpost of this series, we will discuss education and knowledge sharing. We will tell you why it is important that everyone at your organisation is familiar with the possibilities of Artificial Intelligence / Machine learning and why you should keep your team of experts knowledgeable about the newest possibilities in this field.

Let's get in touch